基于改进LOG算子与Otsu算法结合的物体表面图像残损检测方法

针对物体表面残损区域存在明显的亮缺陷和不明显的暗缺陷这一特性,构建一种基于改进LOG算子与Otsu算法相结合的物体表面残损区域缺陷的边缘检测方法。首先,针对传统的LOG算子在检测图像亮缺陷边缘时检测结果不准确的问题,引入可根据图像特性自动调整模糊因子的Wiener滤波代替传统LOG算子中的高斯滤波,以提高图像亮缺陷检测的精度;其次,针对检测图像暗缺陷边缘时结果不准确的问题,使用Otsu算法分析图像暗缺陷的灰度直方图来自动确定阈值,以提升暗缺陷边缘检测准确率;最后,采用像素加权平均融合算法对检测出的物体表面图像亮、暗缺陷边缘进行融合,以实现物体表面残损缺陷检测。实验结果表明:相较于单独使用改进的LOG算子和Otsu算法,采用加权融合的方法检测到的缺陷像素点数量与原始图片中基本一致,能够更准确地对图像中物体表面残损区域的亮、暗缺陷边缘进行检测。

HIGH RESOLUTION PARAMETRIC MODELLING FOR TWO-DIMENSIONAL RADAR TARGET USING PRONY ALGORITHM

On the conditions of low-resolution radar, a parametric model for two-dimensional radar target is described here according to the theory of electromagnetic scattering and the geometrical theory of diffraction. A high resolution estimation algorithm to extract the model parameters is also developed by building the relation of the scattering model and Prony model. The analysis of Cramer-Rao bound and simulation show that the method here has better statistical performance. The simulated analysis also indicates that the accurate extraction of the diffraction coefficient of scattering center is restricted by signal to noise ratio, radar center frequency and radar bandwidth.

Two-dimensional cross entropy multi-threshold image segmentation based on improved BBO algorithm

In order to improve the global search ability of biogeography-based optimization(BBO)algorithm in multi-threshold image segmentation,a multi-threshold image segmentation based on improved BBO algorithm is proposed.When using BBO algorithm to optimize threshold,firstly,the elitist selection operator is used to retain the optimal set of solutions.Secondly,a migration strategy based on fusion of good solution and pending solution is introduced to reduce premature convergence and invalid migration of traditional migration operations.Thirdly,to reduce the blindness of traditional mutation operations,a mutation operation through binary computation is created.Then,it is applied to the multi-threshold image segmentation of two-dimensional cross entropy.Finally,this method is used to segment the typical image and compared with two-dimensional multi-threshold segmentation based on particle swarm optimization algorithm and the two-dimensional multi-threshold image segmentation based on standard BBO algorithm.The experimental results show that the method has good convergence stability,it can effectively shorten the time of iteration,and the optimization performance is better than the standard BBO algorithm.

Design of a Two-dimensional Recursive Filter Using the Bees Algorithm

This paper presents the first application of the bees algorithm to the optimisation of parameters of a two-dimensional （2D） recursive digital filter. The algorithm employs a search technique inspired by the foraging behaviour of honey bees. The results obtained show clear improvement compared to those produced by the widely adopted genetic algorithm （GA）.

改进沙猫群优化算法的2D-OTSU多阈值图像分割

针对传统多级阈值图像分割方法精度低、收敛速度慢的问题,提出一种改进的沙猫群优化算法(Improved Sand Cat Swarm Optimization, ISCSO)用于全局优化,并应用于2D-OTSU多阈值图像分割任务。通过使用Henon混沌映射和反向折射机制初始化种群,使得种群的分布更加均匀,提高搜索的起始状态,从而增加算法的全局搜索能力;采用非线性灵敏度更新公式来平衡搜索多样性和收敛精度;引入可变螺旋搜索策略改进位置更新算法,以确保算法具有较好的搜索多样性和跳出局部最优解的能力。选取6张测试图像对ISCSO算法进行2DOTSU多阈值图像分割实验,采用峰值信噪比(PSNR)、特征相似性指数(FSIM)和结构相似性指数(SSIM)对实验结果进行评价。实验结果表明,基于ISCSO算法的2D-OSTU多阈值图像分割方法在图像分割任务中85.2%的结果优于对比算法,具有较强的搜索精度和收敛速度,这证明了ISCSO算法在图像分割领域的有效性和潜力。

A Two-dimensional Genetic Algorithm Based on the Eno-Haar Wavelet Transform

A two-dimensional genetic algorithm of wavelet coefficient is presented by using the ENO wavelet transform and the decomposed characterization of the two-dimensional Haar wavelet. And simulated by the ENO interpolation the article shows the affectivity and the superiority of this algorithm.

Dynamic threshold for SPWVD parameter estimation based on Otsu algorithm

Time-frequency-based methods are proven to be effective for parameter estimation of linear frequency modulation （LFM） signals. The smoothed pseudo Winger-Ville distribution （SPWVD） is used for the parameter estimation of multi-LFM signals, and a method of the SPWVD binarization by a dynamic threshold based on the Otsu algorithm is proposed. The proposed method is effective in the demand for the estimation of different parameters and the unknown signal-to-noise ratio （SNR） circumstance. The performance of this method is confirmed by numerical simulation.

Image Segmentation of Brain MR Images Using Otsu’s Based Hybrid WCMFO Algorithm

In this study,a novel hybrid Water Cycle Moth-Flame Optimization(WCMFO)algorithm is proposed for multilevel thresholding brain image segmentation in Magnetic Resonance(MR)image slices.WCMFO constitutes a hybrid between the two techniques,comprising the water cycle and moth-flame optimization algorithms.The optimal thresholds are obtained by maximizing the between class variance(Otsu’s function)of the image.To test the performance of threshold searching process,the proposed algorithm has been evaluated on standard benchmark of ten axial T2-weighted brain MR images for image segmentation.The experimental outcomes infer that it produces better optimal threshold values at a greater and quicker convergence rate.In contrast to other state-of-the-art methods,namely Adaptive Wind Driven Optimization(AWDO),Adaptive Bacterial Foraging(ABF)and Particle Swarm Optimization(PSO),the proposed algorithm has been found to be better at producing the best objective function,Peak Signal-to-Noise Ratio(PSNR),Standard Deviation(STD)and lower computational time values.Further,it was observed thatthe segmented image gives greater detail when the threshold level increases.Moreover,the statistical test result confirms that the best and mean values are almost zero and the average difference between best and mean value 1.86 is obtained through the 30 executions of the proposed algorithm.Thus,these images will lead to better segments of gray,white and cerebrospinal fluid that enable better clinical choices and diagnoses using a proposed algorithm.

Vertical two-dimensional non-hydrostatic pressure model with single layer

The vertical two-dimensional non-hydrostatic pressure models with multiple layers can make prediction more accurate than those obtained by the hydrostatic pres- sure assumption. However, they are time-consuming and unstable, which makes them unsuitable for wider application. In this study, an efficient model with a single layer is developed. Decomposing the pressure into the hydrostatic and dynamic components and integrating the x-momentum equation from the bottom to the free surface can yield a horizontal momentum equation, in which the terms relevant to the dynamic pressure are discretized semi-implicitly. The convective terms in the vertical momentum equation are ignored, and the rest of the equation is approximated with the Keller-box scheme. The velocities expressed as the unknown dynamic pressure are substituted into the continuity equation, resulting in a tri-diagonal linear system solved by the Thomas algorithm. The validation of solitary and sinusoidal waves indicates that the present model can provide comparable results to the models with multiple layers but at much lower computation cost.

基于分块Otsu法的车窗开关缺陷阈值分割方法

车窗开关的外观与性能有着不可分割的关系,表面缺陷检测是阻止缺陷产品流入市场的重要手段。针对用传统的阈值算法对车窗开关进行分割时容易造成误分现象,提出一种分块Otsu阈值法。对开关组图像去噪后分割开关,然后对开关图像进行分块阈值处理,通过提取特征进行缺陷判断。分块阈值处理是将图像分块,以整幅图像灰度最高值作为基准,对每块图像进行灰度补偿,并进行Otsu阈值法处理。实验结果表明:分块Otsu法能有效分割出车窗开关的缺陷,同时消除了光照不均现象的影响,充分验证该方法的可行性;利用多种常见阈值分割处理进行对比实验,其效果优于其他处理方法,充分验证方法的高效性。

The Study of Otsu Algorithm Applied in the Measuring of Ash Proportion

To further improve the boiler ash ratio detection methods and resource utilization, through image processing technology for boiler ash ratio analysis, the article first studied the one-dimensional Otsu algorithm, and then for the one-dimensional Otsu algorithm, in order to improve the accuracy of the algorithm, then it puts forward a two-dimensional Otsu algorithm. Finally the two-dimensional Otsu algorithm combined with the one-dimensional Otsu algorithm and the improved Otsu algorithm. By analyzing the improved Otsu algorithm, this paper considers the pixel gray value, neighborhood information, excluding light, noise and the relative efficiency of one-dimensional Otsu algorithm higher accuracy. The relative dimensional Otsu algorithm operating efficiency has been greatly improved. Improved Otsu algorithm in dealing with boiler ash ratio detection has played a very good part in the ecological environment, economic development and some other important aspects.

改进蜜獾算法优化OTSU的图像分割研究

群体智能算法结合图像分割技术已经成为图像处理领域中的新热点,传统的图像分割方法需要大量的人力和时间,蜜獾算法(honey badger algorithm,HBA)可以通过模拟蜜獾觅食的行为来执行优化任务,在寻找解决问题的过程中可以逐步逼近最优解来实现图像分割任务;通过反向学习策略改进蜜獾种群的初始化,提高种群多样性和分布平衡,从而提高算法的整体搜索能力;引入柯西变异因子,对算法计算得到的可行解进行扰动,使算法更易于跳出局部最优,增强算法的局部搜索能力和收敛精度;选取三幅测试图像进行分割验证,实验结果显示,融合改进蜜獾算法和二维OTSU算法得到的分割图像精度更高、效果更细致,验证了方法的有效性;综上所述,改进蜜獾算法具有更好的鲁棒性和泛化性,优化的二维OTSU算法可以更好地处理复杂场景和图像。

基于Otsu的EEG通道选择情绪识别研究

脑电信号情绪识别是数据人机交互(HCI)技术的一种,实时情感识别对于模型性能要求较高,为实现以较低的运算成本获取较高的识别精度,采用时域滑动窗口的方法扩充样本量,基于Otsu算法筛选出含有最多情绪特征信息的通道,并利用快速傅里叶变换进行脑电信号频段提取,以功率谱密度作为特征,构建了基于支持向量机等分类模型,对高唤醒-低唤醒(HA-LA)和高效价-低效价(HV-LV)两种任务进行分类。实验表明,使用SVM分类器在HA-LA情绪识别任务中得到(82.2±0.4)%的识别准确率,在HV-LV情绪识别任务中得到(83.4±0.3)%的识别准确率。所提出的时域滑动窗口能有效提取含有情绪的脑电信号,在减少数据量的情况下仍获得了不错的情绪识别性能,为实时情感识别的脑机接口提供了一种高效的模型。

基于GA-PSO-Otsu算法的质子交换膜燃料电池催化层孔结构自适应识别

车载质子交换膜燃料电池催化层的孔结构识别效率低、精度差且实验要求严格,无法适应日趋规模化的行业发展体系,因此针对该问题,本文提出基于遗传粒子群的最大化类间方差(GA-PSO-Otsu)优化算法,实现对催化层扫描电镜图孔径分布和孔隙率高效、精确且自适应的识别和测算。首先,协同引入高斯卷积核与二值化阈值最大化类间方差,有效降低噪声和手动调参对精度和效率的影响,实现自动化去噪和孔结构识别;其次,进一步提出遗传粒子群算法,有效解决传统方法遍历参数耗时长和易陷入局部优化的问题,兼具高精度和高效率的优点;最后,通过对催化层结构和灰度分布差异明显的扫描电镜图的对比实验验证,表明该方法具备良好的鲁棒性、自适应性和实用性,与遍历所有参数的传统Otsu算法的孔隙率误差小于0.5%,测算耗时降低约26.2%。

基于OTSU图像分割算法的碎米检测

碎米作为大米加工过程的常见产物,常会对产品的口感、味道产生影响,因此针对整米中碎米的有效筛分尤为重要。针对上述问题,该文建立基于大津法(maximal variance between clusters,OTSU)图像分割算法的逻辑回归模型用以检测整米中的碎米。将检测结果与国标法进行对比,结果表明逻辑回归模型的曲线线下面积(area under the curve,AUC)值为0.987,柯尔莫可洛夫-斯米洛夫(Kolmogorov-Smirnov,KS)值为0.909,0.5为最佳阈值;而国标法的AUC值为0.922,KS值为0.669,21为最佳阈值。该文所建立的逻辑回归模型的准确率、精确率、召回率及F1分数均高于国标法。此外,逻辑回归模型的AUC值比国标法的AUC值更接近于1,KS值也更高,表明逻辑回归模型能够更好地区分碎米与整米。长轴(x_(1))、面积(x_(2))、短轴(x_(3))与长短轴比(x_(4))4个特征参数都是模型中具有显著影响的因素,对应的线性关系为z=-139.97-5.35x_(1)+10.93x_(2)+2.86x_(3)+34.59x_(4)。

非接触电导检测土壤养分离子的谱峰自动识别方法

[目的/意义]电容耦合非接触式电导检测(Capacitively Coupled Contactless Conductivity Detection,C4D)在农业土壤养分离子检测方面发挥着重要作用。对C4D信号中离子特征峰的有效识别,有利于后续对离子特征峰的定性和定量分析,为加强农业土壤养分管理提供依据。然而,C4D信号的特征峰检测仍然存在无法自动精准识别、人工操作复杂、效率低等缺点。[方法]提出一种基于连续小波变换结合粒子群优化(Particle Swarm Optimization,PSO)和最大类间方差法(Otsu)的谱峰自动识别算法,旨在实现准确、高效、自动化的C4D信号峰识别。采用C4D检测样品溶液,得到离子谱图信号,对谱图信号进行连续小波变换,得到小波变换系数矩阵。通过搜索小波系数变换系数矩阵极值,识别出脊线和谷线。将小波系数矩阵转换为灰度图像,结合PSO和Otsu寻找最佳阈值,进一步对灰度图像的背景和目标分割,再结合原始谱图中的脊谷线识别谱图中的特征峰。[结果与讨论]测试含有41、61和102个峰的数据集,以受试者工作特性(Receiver Operating Characteristic,ROC)曲线和度量值作为评估峰值检测算法性能的准则。与其他方法相比,基于连续小波变换结合粒子群优化的最大类间方差法分割图像(Continuous Wavelet Transform C.ombined with Particle Swarm Optimization of Otsu to Segment Image,CWTSPSO)的谱峰自动识别算法的ROC曲线均保持在0.9以上,度量值分别为0.976、0.915和0.969。CWTSPSO能够有效检测出更多弱峰和重叠峰,同时检测出更少的假峰,有利于提升C4D信号的谱峰识别率和精准性。[结论]本研究提出的CWTSPSO能为非接触式电导检测农业土壤养分离子信号分析提供有力支持。

改进多方向Sobel算子的剩余油边缘检测方法

为进一步提高剩余油边缘检测的质量和精度,提出一种改进多方向Sobel算子的剩余油边缘检测方法。该方法通过改进双边滤波方法去除微观剩余油分布图像噪声,实现保边去噪的目的;结合Otsu算法使其能自适应获得微观剩余油图像的最佳阈值;在改进的4个方向上依次使用幅度为3×3的Sobel算子计算剩余油图像梯度的幅值和方向;最后利用非极大值抑制算法过滤伪边缘像素点,从而得到最终的剩余油边缘检测图像。实验结果表明,该方法能在去除微观剩余油图像噪声的同时,精确检测出微观剩余油分布图像中的剩余油边缘信息。

基于加权灰度图与混合阈值分割方法的光伏热斑检测

在光伏红外热图像中,热斑和部分高温工作区的亮度非常接近。在利用传统的阈值分割技术提取热斑时,往往会将工作区也一并分割出来,形成虚假热斑。结合Otsu算法和Sauvola算法的优点,提出了一种基于加权灰度图的混合阈值分割方法。通过对灰度图加权处理,降低工作区亮度,从而增强热斑与工作区的对比度,改善热斑的可视性。利用Otsu算法与Sauvola算法二值化灰度图,并根据二值图差异度计算混合阈值,以此消除虚假热斑的干扰。实验证明,该方法适用于检测存在高温工作区的光伏板热斑,能够精准有效地分割热斑。

基于改进Canny算法的皮革铺展图像边缘检测研究

为提高皮革抓取铺展过程的准确性,提出一种基于改进Canny算法的皮革铺展图像边缘检测方法。该方法首先在皮革铺展图像分割中采用二维Otsu算法,并引入狼群算法对二维Otsu算法阈值进行优化,以改进传统的Canny边缘检测阈值由人为设定的问题;利用改进后的Canny算法对皮革铺展图像进行边缘检测,从而为后续的皮革抓取铺展奠定基础。结果表明,引入狼群算法优化后的二维Otsu算法,其迭代次数与运行时间都相较于原始Otsu算法有了大幅降低,结构一致性指数提高了0.024,具有更好的图像处理效果;使用改进的Canny算法,相较于原始Canny算法以及Robert、Prewitts等常用边缘检测算法,不仅实现了阈值的自适应设置,得到的皮革边缘线条也更完整清晰,具有更好的抗噪性。将改进Canny算法用于皮革双机械臂抓取铺展,机械臂可准确检测皮革铺展图像的边缘。由此得出,本方法可为后续的铺展抓取奠定基础,具有一定的实用性。

基于改进Canny算法和Hough变换的脉冲氩弧焊坡口边缘提取研究

针对脉冲氩弧焊中被动光传感采集的图像存在光亮变换快,目标像素少,难以提取出坡口的问题,提出一种基于改进阈值选择和双阈值检测的Canny算法。通过中值滤波和双边滤波代替传统Canny算法中的高斯滤波对图像进行滤波,引入背景像素与前景像素的数量比改进Otsu算法的最大方差,计算得到了靠近坡口的梯度值的阈值,依据改进后的双阈值检测获得完整的坡口边缘信息,对边缘信息进行Hough变换提取出清晰的坡口边缘。实验结果表明,和传统Canny算法相比,改进后的Canny算法提取坡口成功率提高,获得的坡口边缘信息也更加完整。